Welcome back to another season of counterintuitive football betting! Last season, after Gamblor burrowed back under the seabed for its annual six-month hibernation, I took a look at the Martingale strategy of doubling your bet after every loss and concluded that everyone who has ever written about in the past is absolutely right – it really is a terrible strategy that will never work in practice. And although Gamblor was unsuccessful for the season (dropping a total of $353 for the year, mostly due to the Green Bay Packers and the Seattle Seahawks and bringing its lifetime total down into negative territory), it actually was a profitable season, in an oblique way. During Week 14 last year, I looked at all the stocks whose ticker symbols matched the three-letter notations I use in Gamblor’s code. I decided to give one of them a shot – SEA – and it ended up going up 20% in six months before the fund was liquidated because its lazy shareholders (including yours truly) didn’t send in their proxy votes and they couldn’t get a quorum for a new investment strategy.
Near the end of last season I decided to tinker with Gamblor a little bit to see if I could squeeze a few more percentage points out of it, and to eliminate the clumsy structure of confidence tiers that it relies upon. And so now, I’d like you to raise your glasses and drink a toast to the proud father and his unholy spawn. I call him SON OF GAMBLOR!
The original Gamblor worked by looking back at between 9 and 20 weeks worth of previous games and determining a series of theoretical score pairs for the upcoming week. For each of these simulations (501 theoretical score pairs for each game for each of 360 different models), the computer added in the spread and calculated the percentage it believed a bet on each team would win (referred to as “confidence level”). If it reached a confidence level of 85% or greater, it would put a bet down – on the opposite team. For confidence levels of greater than 85%, it would put down 3 units, for confidence levels of greater than 90%, it would put down 8 units, and for confidence levels of 95% or greater it would put down 25 units. For games where it thought the spread was perfect, it would put down a hedge bet of 1 unit on the underdog.
Last year Gamblor performed according to theory, which is to say the losses it delivered last year were perfectly normal, if regrettable. However, Gamblor tends to favor underdogs. That might be a pretty sound strategy (since underdogs have a slight edge over the long term) but it’s not a whole lot of fun, and it’s hell on your nerves to rely on two or three backdoor covers* every week. I wanted to bet on favorites more often and have the chance to sit back and enjoy those runaway blowouts against Kansas City and the St. Louis Rams. So I added two new twists. The first is to determine the distance a team has to travel to play the game, and to subtract an amount from their projected score based on this distance (i.e. the farther a team travels, the greater the penalty). The second is to give a team a bonus or penalty based on the time change they experienced in traveling. Teams that travel east, where they’re forced to wake up earlier than they’re used to, are subject to a small penalty. Teams that travel west get a small bonus. This skewed things towards favorites a little bit, and actually improved the program’s performance by a couple of percentage points.
In addition to the new factors in the model, I changed the betting scheme to break away from the tiered structure, which doesn’t distinguish between confidence levels within the 5-point range – Gamblor bets the same number of units whether it’s 90.5% confident or 94.5% confident. The betting scheme for Son of Gamblor consists of a theoretical curve described by the following equation:
This is actually based on Son of Gamblor’s winning percentage over the last thirteen years. While you can see in the graph above that it doesn’t exactly match the actual winning percentage, it’s pretty close. But that’s by design – it means that the meat of Son of Gamblor’s winnings come from the 86% to 92% range, which are relatively common (Son of Gamblor picks about 6 or 7 games every week) and means it won’t be relying on the hedge bets that made up so much of the action last year with the original Gamblor. Son of Gamblor doesn’t actually make any hedge bets at all. The number of units to bet is simple – it’s the theoretical percentage minus fifty. Thus for a particular model (remember, there are 360 models in all) where Son of Gambor is 90% certain of it’s pick, you’d bet 61 – 50 = 11 units. Here’s how it compares to the original Gamblor over the long term – you can see that he’s not nearly as volatile as his temperamental father, so we’ll be able to get accustomed to a smoother ride.
This second graph shows where the money comes from – most of the actual cash winnings are (ahem, I mean ‘should be’) delivered by bets in the ranges from 86% to 92%. Bets of less than 86% confidence aren’t correct often enough, and bets of greater than 92% confidence are just too rare.
Anyhow, let’s move on to the picks. I haven’t figured out how to format things appropriately so you can see Son of Gamblor’s picks, so I’ve put his opinion in bold down in the comments section for each game. Hopefully by next week I’ll have a simple table that summarizes things a little more succinctly, along with my own picks (which are usually terrible). Good luck!
*A backdoor cover is when the favorite runs out to a big lead early, but the underdog picks up enough meaningless points at the end of the game to cover the spread.
NEW ORLEANS SAINTS
Gamblor’s Pick: Minnesota (+5)
Bet Amount: $13.15 (Hedge)
|COMMENTS: Sure, the Saints are the Superbowl champions, but the previous matchup between these teams couldn’t have been closer. Horrific play calling by Childress and Farve’s interception ruined what should have been an away victory for the Vikings. This ought to be a close game – five points is too many. I like what Gamblor is doing here.|
NEW YORK GIANTS
Gamblor’s Pick: NY Giants (-7)
Bet Amount: $226.60 – Flip (95%)
|COMMENTS: Is this line some kind of joke? The last time these two teams played – in the second-to-last game of the 2009 season – the Panthers thoroughly humiliated the Giants, beating them 41-9 and knocking them out of playoff contention. How much could have possibly changed during the offseason? Have the Giants gotten that much better? Have the Panthers gotten that much worse? Have the Giants gotten any better? While the Panthers have struggled mightily during preseason, the Giants – in particularly their offense – haven’t really done anything to warrant this kind of optimism.
It’s situations like this where Gamblor (theoretically) thrives. A spread of seven points seems absolutely absurd for a pair of teams that are passing each other on the downslope towards mediocrity. Why such a high line? Gamblor’s theory is that the books are expecting the Giants to win this game handily, and thus want to try to discourage bettors from wagering on the host.
Son of Gamblor agree wholeheartedly with his dad – he’s coming out swinging with a bet of $242.21 on the host.
Gamblor’s Pick: Miami (-3)
Bet Amount: $3.08 (Flip)
|COMMENTS: Too little money for a game that nobody in their right mind will care about. Son of Gamblor also likes Miami, but only for $2.06. I guess since the combination exceeds $5 and also I’m a degenerate gambler that would set an over/under for how many homeless people I’ll walk by on the way to work (this morning I set it at 5, and actually passed zero – guess they slept late), I’ll put a bet in, but I’m pretty indifferent to this contest.|
Gamblor’s Pick: Pittsburgh (+2.5)
Bet Amount: $8.01 (Hedge)
|COMMENTS: Atlanta looked very much out of synch during the preseason, and Pittsburgh is going to come out swinging. I’m happy to be getting the points here, and I actually think there’s a good chance Pittsburgh will win outright.|
Gamblor’s Pick: Detroit (+6.5)
Bet Amount: $24.04 (Flip)
|COMMENTS: This one feels like a gift. Between an improving Stafford, Calvin Johnson, and rookie Jahvid Best, this Lions team has the potential to explode at times on offense. They’ll still be chained to the corpse of their defense, but against Chicago that won’t be as big of a liability as when they face their other NFC North opponents Green Bay and Minnesota. I wouldn’t be surprised to see Forte have a big game, but Cutler is still trapped behind an offensive line that provides about as much physical obstruction to opponents as a room full of ghosts. Son of Gamblor also really likes the Lions as well here, wagering $39.99 on Detroit.|
NEW ENGLAND PATRIOTS
Gamblor’s Pick: Cincinnati (+4)
Bet Amount: None
|COMMENTS: Gamblor thinks this line is just a shade lower than it should be, but not enough to suspect a trap. Personally, I think the Patriots are going to hit the ground running and win this game handily.|
TAMPA BAY BUCCANEERS
Gamblor’s Pick: Cleveland (+3)
Bet Amount: $3.90 (Hedge)
|COMMENTS: In their first disagreement of the season, Gamblor likes Cleveland as a small hedge bet, but Son of Gamblor thinks the travel will wear out Cleveland enough to give Tampa Bay an edge here. Son of Gamblor wants to wager $13.86 on Tampa, so at the end of the day I’ll be putting down $9.96 on the Buccaneers. I’m fine with that – I’ll be rooting for them anyhow.|
Gamblor’s Pick: Denver (+2.5)
Bet Amount: $14.38 (Hedge)
|COMMENTS: Kyle Orton has looked pretty sharp this preseason, and while I know that it’s silly to get excited about preseason results, I’ll be damned if I’m not optimistic for the Broncos’ chances in this game. Jacksonville has a reputation of phoning in any and all games against non-divisional opponents, and they’ll be playing in front of stands full of fans that will be cheering for one of their opponents players. The Jaguars’ ownership decided to pass on Tim Tebow during the draft, which makes me suspicious that they’re on a Major League type crusade to escape Jacksonville and get themselves a brand new stadium in Los Angeles in the near future. Son of Gamblor doesn’t like how much travel Denver’s going to be doing or the time change they’ll be facing, so he’s supporting Jacksonville to the tune of $0.93.|
Gamblor’s Pick: Houston (+3)
Bet Amount: $29.99 (Hedge)
|COMMENTS: There’s a couple of factors that should give me comfort here – Superbowl losers tend to come out flat in their first game the following season, and this is one of the biggest games in the history of a very hungry and steadily improving Houston franchise. Knowing these things doesn’t help, though – Peyton Manning just keeps finding ways to win week after week. At the very least I’ll be hoping for a shootout to keep things interesting, but I have a hard time seeing the Colts lose this game, even on the road.|
Gamblor’s Pick: Oakland (+6)
Bet Amount: None
|COMMENTS: I’m glad there’s no money on the line here. I have absolutely no idea how well Oakland and Tennessee will play this year. Chris Johnson won’t be sneaking up on the Raiders defense, but that doesn’t mean they’ll be able to stop him And given how well Oakland managed to play once they finally benched Jamarcus Russell, they seem to have a lot of potential. It should be interesting to see what Jason Campbell can do.|
|GREEN BAY PACKERS
Gamblor’s Pick: Philadelphia (+3)
Bet Amount: $25.68 (Hedge)
|COMMENTS: I feel like Gamblor is throwing money away here. Can anyone stop Aaron Rodgers? I’m hoping that this doesn’t turn into another season where the computer pisses away money betting against Green Bay every week. On the other hand, at least Vegas will be losing money at the same time, because at the moment 96% of wagers are on the Packers.|
|SAN FRANCISCO 49ERS
Gamblor’s Pick: Seattle (+3)
Bet Amount: $31.84 (Hedge)
|COMMENTS: Ugh. I really, really, really hate having to make this bet. Seattle, one of Gamblor’s nemeses last year, is once again f***ing terrible, and is once again the recipient of a large hedge bet courtesy of the computer. As you may or may not know, I personally have absolutely no discretion over who the computer bets on, but if I did, you can rest assured and I’d flip this bet.|
SAINT LOUIS RAMS
Gamblor’s Pick: St. Louis (+4)
Bet Amount: $5.55 (Hedge)
|COMMENTS: Once again, it feels like I’m throwing money away here. It’s only $5, but I’m not expecting to see any improvement out of St. Louis this year, and Arizona hasn’t fallen far enough that they won’t be able to run up the score against their inferior opponents. To my dismay, Son of Gamblor is also in support of St. Louis for $7.44.|
Gamblor’s Pick: Washington (+3.5)
Bet Amount: $10.68 (Hedge and Flip)
|COMMENTS: Dallas hasn’t really meshed during the preseason, and I think this game will be symptomatic of a slow start to the season for them. I think Washington will somehow manage to keep things close enough for the spread to factor in, so I’m content with Gamblor’s choice here. Son of Gamblor is a big Redskins fan in this game, wagering $40.36 on the host.|
NEW YORK JETS
Gamblor’s Pick: Baltimore (+2.5)
Bet Amount: $39.44
|COMMENTS: It’s hard to envision the Jets losing their first game of the season after such a strong finish last year, but their offense has been nonexistent during the preseason. With Revis back, you can expect Baltimore to have trouble moving the ball through the air, and I don’t know that they’ll have all that much luck on the ground either. Expect Ray Rice to receive a lot of touches in this game, expect a lot of field goals, and expect a pair of final scores that are only different by a few points. This one is a coin toss for me, so in the spirit of solidarity I’ll get Gamblor’s back and go with the Ravens on this one.|
|SAN DIEGO CHARGERS
KANSAS CITY CHIEFS
Gamblor’s Pick: San Diego (-4.5)
Bet Amount: None
|COMMENTS: I’m a bit surprised the spread isn’t bigger here. San Diego looks to pick up where they left off last season, except with a revamped running game thanks to the addition of Ryan Matthews. Kansas City looks like they’ll be plagued by the same inept coaching decisions and talentless personnel. San Diego should walk away with an easy victory here. This actually turns out to be Son of Gamblor second-biggest bet of the weekend, as it throws down $72.28 in support of the hapless Chiefs. I’ll be hoping for a miracle in this one, though apparently Vegas will as well as 92% of bets are on the Chargers thus far.|